Motivation and emotion/Book/2015/Emotion elicition using film

Overview
Emotions play a crucial role in normal and abnormal human functioning. Psychologists have long been eager to study its role in brain activity under reliable and reproducible laboratory study. Before film many methods such as hypnosis or threats were used that crossed ethical boundaries and were unreliable in their results (Gross & Levenson, 1995). From its first study in 1962 the use of motion film has advances in leaps and bounds to prove itself the most reliable and popular way to elicit emotions in a laboratory setting (Coan & Allen, 2007). This book chapter explores the history of film and its use in emotion elicitation, what emotions are and how they can be measured, how emotions are elicited in the laboratory and finally how films are used to elicit emotions as well as which films are recommended to use for specific emotions.



Early Film
Films began their run in history in the 1890s with the introduction of motion picture cameras, at this time films had no sound and were limited to times under a minute. It could be said that the first emotions ever elicited by film were that of amazement at the spectacle of moving pictures like people had never seen before (Bordwell & Thompson, 2003). Over the next three decades of silent films the technology moved from a novelty to an entertainment industry which gained extreme popularity. The year 1900 marks the point where motion picture transformed into film and film-makers would use actors and editing techniques to tell stories designed to provoke emotions and entertain their audience (Bordwell & Thompson, 2003). From this point on films became more extravagant, and the industry reached audiences all around the world making cinemas a social hub in cities around the world, the age of films eliciting emotions to audiences of people had begun its spread around the world (Bordwell & Thompson, 2003).

Modern History
Since its first laboratory use found success the use of films in emotional elicitation has established a long history of studies. Early studies focused on simply producing high intensity states of emotional arousal such as Goodenogh’s 1975 study on the effects stressful films can have on individuals once they go to sleep or Pillard’s study into the effects of marijuana on experimentally induced anxiety (Goodenough, Witkin, Koulack, & Cohen, 1975; Pillard, McNair & Fisher, 1974). Studies around this time simply used Lazarus’ proven method to produce the stress response they needed for their studies following the same method (Gross & Levenson, 1995). As time went on, research lead to more specific emotions such as fear or sadness being needed for study and work was carried out, W.A Brown was the first to study how the specific emotion Anger could be induced in a laboratory with film (Brown, Corriveau & Monti 1977; Gross & Levenson, 1995). Later studies have been conducted on the use of cuts of feature films being used to elicit up to 6 emotional states: amusement, anger, disgust, fear, neutral and sadness. Philippot (1993) set the new standard for the variety of emotions that could be induced in a laboratory setting using cuts from popular Hollywood films (Philippot, 1993; Gross & Levenson, 1995). Today film clips are utilized to elicit a variety of emotions in a variety of research studies. Researchers use segments from feature length films to aid in studies of emotion and mood for all ages, genders, cultures and more because of the vast variety of studies on the effectiveness of current films to produce such emotions in a diverse list of participants (Gabert-Quillen, Bartolini, Abravanel & Sanislow, 2015).

Emotions


Emotion is a brief, subjective and conscious experience formed primarily by psychophysiological expression, biological reactions and mental states. It is believed to be influenced by mood, temperament, personality, disposition and motivation (Reisenzein, 2007). Its functions include: (Trappl, 2002)


 * 1) The elicitation of Automatic responses and endocrine responses (release of hormones, rising of heart rate). Preparing the body for action
 * 2) Reinforce or punish behaviour through stimulus-reinforcement or stimulus-punishment loop to assist in learning
 * 3) Emotion is motivating- Motivates us to act in certain ways through our learned emotional responses.
 * 4) Communication. Through facial features, sweating etc... Emotion can communicate feelings to others with its physiological responses.
 * 5) Social Bonding- Emotions can produce bonds between family and friends

Measuring Emotions
Measuring these emotions in a laboratory can prove difficult. Measuring all the expressive language of a person, their physiological statistics and observed behaviours could imply an emotion for that person when none is present, this may be the way this person always is. To effectively measure the emotions of individuals their data must be compared and analysed to their resting measurements ({{grammar{{ Their affective state prior to film exposure) to highlight spikes, peaks and anomalies and compare them to what is known about how people respond when expressing emotions (Coan & Allen, 2007). To establish resting states, researchers show a neutral-emotion-film and measure emotional factors while watching this. In addition to providing a resting baseline it also provides the measurements of individuals when they are simply viewing any kind of dynamic external stimuli such as a film (Coan & Allen, 2007). This eliminates and confounding effects of simply watching any film that could alter the results of the research.

Actual measurements inferring the emotional state of participants include:

Self-Reported emotional arousal

This involves participants answering questions in regard to their emotional state at a variety of times around watching the films. The intensity of their subjective emotional states is generally assessed using a sort of scale such as a 7-point scale (Schaefer, Nils, Sanchez & Philippot, 2010). Participants are asked questions such as “How did you feel while watching the film?”, and they would answer on a scale of 1-7 with 1 being “I felt no emotions at all”, and 7 being “I felt very intense emotions”. This form of measurement is cheap and easy to do however is not always accurate because it relies on participants reported information which can be altered by a number of issues such as their understanding of the question, response bias, rating scales or introspective ability (Austin, Gibson, Deary, McGregor & Dent, 1998).

Differential Emotions Scale

To test for discrete emotions felt researchers use a Differential Emotions Scale (DES; Schaefer et al., 2010). This test involves showing participants groups of emotional adjectives such as (1) Interested, concentrated, alert; (2) Sad, downhearted, blue; (3) Warm hearted, gleeful, elate. As well as the adjectives outlined in the DES model researchers can add their own adjective groups to better cater the test towards their study’s needs (Schaefer et al., 2010). For each group of adjectives participants stated how much they felt these emotions while watching the film on a scale such as the 7-point scale (1= “not at all”, 7=”very intense”). This measurement method can gain useful information on a wide range of specific emotions felt while watching a film; however it has all the expected drawbacks mentioned to exist with self-reported information.

Rating Dials

While receiving a stimulus in a laboratory participants place their hands on some kind of dial they can move which represents the feelings they are having of an emotion. The dial can scales from very negative to very positive in regards to any emotion the researchers are trying to measure (Coan & Allen, 2007). This measuring method allows researchers to attain live self-reported feedback from participants that they can accurately pinpoint to a time in the stimuli such as a part of the film the participant was watching (Coan & Allen, 2007).

Facial Action Coding System

This system developed in the 1970s by Ekman and Friesen utilises facial movements from expressions as a unit of measurement called Action Units (Ekman, Friesen & Tomkins, 1971). Specific action unit combinations are known to reflect emotions. By measuring these combinations of action units the facial expressions of people can be accurately used to determine the emotions they are feeling (Ekman, Friesen & Tomkins, 1971).

Startle Eyeblink modulation

This measurement method utilises participant’s involuntary eyeblinks when exposed to a startle to infer their emotional state. It is discussed in full below under the heading Emotional Elicitation, Methods to elicit emotions in the Laboratory.

Skin Conductance Level

Skin Conductance level (SCL) is a commonly used technique in studies to test the responses of the autonomic nervous system. This assessment of the body’s natural response is another factor leading to the inference of emotions in participants (Lazarus et al., 1962). It is commonly measured with a “Dermback” LE-538; this device fits two electrodes on participant’s fingers (Fernández, Pascual, Soler, Elice, Portella & Fernández-Abascal, 2012). A current is run through the finger between these two electrodes and the resistance is measured giving the output of the skin conductance level. This number is altered by several bodily factors including sweat production which is associated with the bodies stress response (Fernández et at., 2012). SCL levels while watching the film can be continually measured and compared to that of the established resting state to discover if there is an emotional response detected at any point throughout the film (Fernández et at., 2012).

Heart rate

Another commonly used technique testing responses from the automatic nervous system is heart rate levels. Heart rate is measured using pulse counters such as a ‘Polar 600’ commonly used in laboratory studies (Fernández et at., 2012). Devices like this place a sensor on participant’s chests which monitors their Heart rate and send the number of beats per minute to a computer. Similarly to the SCL this device can monitor heart rate throughout the entire time of the participants watching the film and give exact data of variations in heart rate, maximum scores and minimum scores and the corresponding moments of the films to infer when a participant may be having an emotional response and to what stimuli (Fernández et at., 2012).

Beyond all self-reported measures and physiological measurements additional data is sometimes attained to fill any inconsistencies in the data and attempt to ensure all confounding factors are accounted for. Demographics such as sex, race, gender or family income are generally attained and other factors such as fainting history, anxiety sensitivity, preference to watching films, disgust sensitivity and more are occasionally factored into final results to try and find possibly overlooked factors regarding emotional arousal through the use of film (Valentiner, Hood & Hawkins, 2005).

Emotional Elicitation
Overview

Human emotion is a complex topic of much research. It is constantly being studied what can induce emotions, what role they have on motivation, their physiological effects, why humans evolved emotions and their place in the grand scheme of consciousness. Whatever the reasons behind studies are experimentalists require reliable means of eliciting emotions under experimental conditions, and they need to elicit these emotions in ethically acceptable ways (Coan & Allen, 2007). In the past emotions such as fear, sadness stress were elicited for research through threats, physical attacks, emotional attacks, severe cold or heat, even exposure to micro-organisms or viruses were used to elicit emotions in participants in the name of scientific discovery (Lazarus et al., 1962). Thankfully today elicitation procedures have advanced to more tightly controlled, replicable and ethical procedures (Coan & Allen, 2007).

Methods to elicit emotions in the Laboratory
Static Photos

A system called the International Affective Picture System (IAPS) is the world’s largest collection of static photos known to elicit emotions. This collection contains over 700 colour photos proven to reliably elicit robust positive, negative or more neutral controlled emotional states in people (Coan & Allen, 2007). All the photos have a standard rating for their arousal level as well as how positive or negative they are (Lang, Ghman & Vaitl, 1988). Another static photo tactic is to display emotional faces with the intent that seeing others emotional faces can elicit the reciprocation of these emotions in participants (Coan & Allen, 2007).

Relived Emotions

Relived emotions involve a semi structured interview aimed at having participants recall and relive emotional life events they have had. By reliving these events as much as possible participants will innately go back to these emotional states and the interview has succeeded in eliciting emotion (Coan & Allen, 2007). These memories can be autobiographical memories or shared memories.

Dyadic Interactions Dyadic interaction involves putting two different people (their relationship is not relevant) together in a comfortable and having them make conversation following a silent baseline period. Topics for conversation can be probed to researchers liking depending on what emotional response they are attempting to elicit (positive, negative; Coan & Allen, 2007). Emotion measurements are taken throughout and following the conversation.

Music and Singing

Like motion films music is an artist’s medium to elicit emotions in their audience as well as entertain. Researchers can have participants listen to music to attempt to elicit this emotional state in the laboratory for their research purposes (Coan & Allen, 2007). Classical music is used to elicit either happy emotions or sad emotions depending on the piece played by the choice of the researcher. Singing can also be utilised to elicit emotions such as humiliation or joy depending on what kind of music participants are asked to sing and the conditions they sing in (e.g. their singing skill, their audience; Coan & Allen, 2007).

Acoustic Startle Reflex

If you make a large enough sound in front of your participant they will act out their innate response found in all humans the Acoustic Startle response. This consists of an automatic initial response of somatic and muscle actions and is followed by a secondary emotional response such as laughing, embarrassment, anger or fear (Soto, Levenson, Ebling, 2005). Once one of these emotions has been elicited researchers can measure what it is they wanted to find out about it.

Startle eyeblink modulation

Startle eyeblink modulation entails attaching measuring devices to a participants face to measure information about their blinking habits when exposed to a startling sound given through headphones. This can give a surprisingly good amount of insight as to a person’s emotional state when exposed to their current stimuli (Coan & Allen, 2007). Research has associated larger eyeblinks when startled with a mix of negative emotions such as anxiety or nervousness. Alternatively smaller eyeblinks are associated with positive emotions such as joy. When coupling this measurement tactic with Film emotion elicitation it can be used to assist in inferring participants’ emotional state Coan & Allen, 2007).

Direct Facial Action Task

Direct Facial action task is where researchers either explicitly direct participants to move their facial muscles to display a certain emotion such as fear or anger, or implicitly give them instructions to move their facial muscles in a particular way without telling them what emotion they will be displaying. While participants are pulling these faces they are asked to give feedback on how they are feeling. It is thought that through making people’s faces show emotions such as anger or joy, they may subliminally start to feel those emotions and thus emotions are elicited through direct facial action tasks (Coan & Allen, 2007).

Eliciting emotion using films
The use of films for emotion elicitation encompasses many of the techniques used above in the one form. Cuts from films can use static photos and moving photos to elicit stronger emotions in participants, they can accompany this with a choice of music to further the effect. While they can’t conduct an interview necessary for relived emotions they can play personal videos of participants in the past or stories similar to what participants have been through to elicit emotions just the same as if they were having an interview recalling those events. They can also utilise methods similar to those of the acoustic startle effect with the audio of their film, they can additionally back this up with images designed to startle (A common trope of the horror genre, the ‘jump scare’). With all of these techniques being encompassed by films it’s no surprise it is the most popular and reliable way of eliciting emotions in the laboratory today (Coan & Allen, 2007).

Methods
Researchers gather their group of participants based on the resources, needs and aims of their study. A large diverse group is usually preferred but not always required or possible. To elicit emotions using films generally participants are isolated (Group viewings are acceptable and sometimes necessary) in any kind of controlled laboratory setting where the researcher can gather measurements with as little confounding factors from the environment possible. Researchers will set up their measuring techniques whether they involve simple self-reported questionaries or full SCL and heart rate equipment. They will have a baseline of measurements gathered whether watching a neutral film or simply before watching anything. To begin the elicitation of emotions through film participants need to be shown a film via a T.V, Cinema screen, hand-held device, any medium possible to show a film to participants is acceptable. Through watching the film emotions should be elicited in the participants and researchers can use data from their measurements to whatever needs of their study.

Which films to induce emotions?
As far as using films that are proven to reliably elicit emotions in participants there has been hundreds of studies on which prove to be the most effective. A recent study “Ratings for emotion film clips” summed up all the films used to elicit emotion in previous study and tested their effectiveness to devise a list of the most effective film clips to elicit emotions through film in a laboratory setting (Gabert-Quillen et al., 2015). This list summarises the best rated clips to elicit the following emotions as found by the study:
 * 1) Amusement: Clips from The Hangover or Modern Times
 * 2) Excitement: Clips from 300 or The Bourne Identity
 * 3) Happiness: Clips from Remember the Titans or Wall-E
 * 4) Calmness: Clips from Pride and Prejudice or Searching for Bobby Fischer
 * 5) Anger: Clips from Crash and Gentlemen’s Agreement
 * 6) Disgust: Clips from National lampoon’s Van Wilder or The Fly
 * 7) Fear: Clips from Psycho or The Ring
 * 8) Sadness: Clips from My Girl or The Shawshank Redemption
 * 9) Surprise: Clips from D.O.A or The Departed

Conclusion
Emotional elicitation is a necessary tool for research on human emotion. The ability to reproduce emotions accurately within a lab setting where confounding factors can be eliminated is essential to the reliability of research conducted. It is clear that since its first use in 1962 film has taken a clear lead as the most effective way to produce emotions under laboratory settings. Hundreds of feature films are produced around the world every year with the intent to elicit emotions and entertain audiences; this is giving researchers rich tools they need to elicit emotions in their labs. Thorough research has been conducted on which films work best at producing emotions and more become available every year. Films are a match made in heaven for the reliable elicitation of emotions in the laboratory.